The purpose of this project was to develop role-based models that leverage linear regression and supervised machine learning to predict the market value of football (soccer) players based on a variety of aspects ranging from player attributes to on-pitch performance statistics. The predictions were then compared to actual market values to analyze how the football market and industry values players of different roles and discuss discrepancies between the model and real-life. The solution was developed using Python’s robust machine learning packages such as scikit-learn and seaborn as well as Jupyter notebook. The data used in this project compiles a list of every player in the top 5 leagues in European football and their performance statistics for the 2019-20 season. The models developed proved effective in determining the market value of footballers given the information found in the dataset and provide insight as to which features are the most valuable in each role on the pitch.